mlsanityData to 14 Oct 2026

Bridging AI and Software Security: A Comparative Vulnerability Assessment of LLM Agent Deployment Paradigms

arXiv 2507.06323 ↗First posted 8 Jul 2025cs.CR cs.AIcited by 7

On the map: LLM agents and tool use → Agent threats, guardrails and governance

Abstract (arXiv)

Large Language Model (LLM) agents face security vulnerabilities spanning AI-specific and traditional software domains, yet current research addresses these separately. This study bridges this gap through comparative evaluation of Function Calling architecture and Model Context Protocol (MCP) deployment paradigms using a unified threat classification framework. We tested 3,250 attack scenarios across seven language models, evaluating simple, composed, and chained attacks targeting both AI-specific threats (prompt injection) and software vulnerabilities (JSON injection, denial-of-service). Function Calling showed higher overall attack success rates (73.5% vs 62.59% for MCP), with greater system-centric vulnerability while MCP exhibited increased LLM-centric exposure. Attack complexity dramatically amplified effectiveness, with chained attacks achieving 91-96% success rates. Counterintuitively, advanced reasoning models demonstrated higher exploitability despite better threat detection. Results demonstrate that architectural choices fundamentally reshape threat landscapes. This work establishes methodological foundations for cross-domain LLM agent security assessment and provides evidence-based guidance for secure deployment. Code and experimental materials are available at https: // github. com/ theconsciouslab-ai/llm-agent-security.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use. Builds on 40 of its 71 references that are in the map. First posted on arXiv 8 Jul 2025.

Builds on

40 of 71 references.

Closest ancestors · 8

Same direction, newest first

  1. 2025Design Patterns for Securing LLM Agents against Prompt Injections10 Jun
  2. 2025TRiSM for Agentic AI: A Review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems4 Jun
  3. 2025A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)4 May
  4. 2025Securing Agentic AI: A Comprehensive Threat Model and Mitigation Framework for Generative AI Agents28 Apr
  5. 2024Swiss Cheese Model for AI Safety: A Taxonomy and Reference Architecture for Multi-Layered Guardrails of Foundation Model Based Agents5 Aug
  6. 2024Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification30 Jul
  7. 2024IsolateGPT: An Execution Isolation Architecture for LLM-Based Agentic Systems8 Mar
  8. 2024InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents5 Mar

Other references · 32

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  4. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
  5. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  6. 2023A Survey on Large Language Model based Autonomous Agentscited by 2,242
  7. 2023Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injectioncited by 1,443
  8. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,420
  9. 2023Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversationscited by 1,207
  10. 2023Safe RLHF: Safe Reinforcement Learning from Human Feedbackcited by 656
  11. 2023A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Uglycited by 387
  12. 2023Formalizing and Benchmarking Prompt Injection Attacks and Defensescited by 364
20 more
  1. 2024PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Modelscited by 267
  2. 2025Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directionscited by 265
  3. 2023Inverse Scaling: When Bigger Isn't Bettercited by 207
  4. 2023An LLM Compiler for Parallel Function Callingcited by 179
  5. 2024AI Agents Under Threat: A Survey of Key Security Challenges and Future Pathwayscited by 127
  6. 2024Automatic and Universal Prompt Injection Attacks against Large Language Modelscited by 119
  7. 2025Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agentscited by 71
  8. 2024TinyAgent: Function Calling at the Edgecited by 58
  9. 2023Identifying and Mitigating Vulnerabilities in LLM-Integrated Applicationscited by 54
  10. 2024Large Language Model for Vulnerability Detection: Emerging Results and Future Directionscited by 45
  11. 2025REAL: Benchmarking Autonomous Agents on Deterministic Simulations of Real Websitescited by 45
  12. 2023A Comprehensive Overview of Backdoor Attacks in Large Language Models within Communication Networkscited by 42
  13. 2025Security Concerns for Large Language Models: A Surveycited by 29
  14. 2024Insights and Current Gaps in Open-Source LLM Vulnerability Scanners: A Comparative Analysiscited by 10
  15. 2024Large Language Model Supply Chain: Open Problems From the Security Perspectivecited by 9
  16. 2025Demystifying LLM Supply Chain Vulnerabilities in the Wild: Distribution, Root Cause, and Real-World Impactcited by 8
  17. 2024A Grading Rubric for AI Safety Frameworkscited by 5
  18. 2024MARCO: Multi-Agent Real-time Chat Orchestrationcited by 5
  19. 2025Breaking the Loop: Detecting and Mitigating Denial-of-Service Vulnerabilities in Large Language Modelscited by 3
  20. 2024Research on WebAssembly Runtimes: A Surveycited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 5.

7 papers.

  1. 2025MCP Security Bench (MSB): Benchmarking Attacks Against Model Context Protocol in LLM Agentscited by 42
  2. 2026Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluationcited by 7
  3. 2025Penetration Testing of Agentic AI: A Comparative Security Analysis Across Models and Frameworkscited by 4
  4. 2026MCP-DPT: A Defense-Placement Taxonomy and Coverage Analysis for Model Context Protocol Securitycited by 2
  5. 2026Large Language Models for Agentic NetOps and AIOps: Architectures, Evaluation, and Safetycited by 2
  6. 2026When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Agentscited by 1
  7. 2026Connecting the Dots in Agentic AI Security: A Cross-Dimensional Threat Taxonomy, Evaluation Maturity, and Open Challengescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025From Prompt Injections to Protocol Exploits: Threats in LLM-Powered AI Agents Workflows
  2. 2026On Understanding, Identifying, and Mitigating Vulnerabilities in Agentic Large Language Models
  3. 2026Breaking the Protocol: Security Analysis of the Model Context Protocol Specification and Prompt Injection Vulnerabilities in Tool-Integrated LLM Agents
  4. 2026LLM-Based Agents for Software and Systems Security: Approaches, Applications, and Assessment
  5. 2025Breaking Agent Backbones: Evaluating the Security of Backbone LLMs in AI Agents
  6. 2026Taming OpenClaw: Security Analysis and Mitigation of Autonomous LLM Agent Threats
  7. 2025Comprehensive Vulnerability Analysis is Necessary for Trustworthy LLM-MAS
  8. 2026The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis